Diastereoselective synthesis of syn-aminoalcohols via contributing CH-π interaction: Simple synthesis of (-)-bestatin

Article abstract of DOI:10.1016/S0040-4039(03)01394-7

¹H NMR and X-ray crystallography studies revealed that a CH-π and chelation control in aromatic aminoaldehydes (1-6) effects a highly diastereoselective addition to afford optically active syn-aminoalcohols (1a-6a). This methodology was applied to the synthesis of (-)-bestatin.

Full text of DOI:10.1016/S0040-4039(03)01394-7

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 44 (2003) 5905–5907
Diastereoselective synthesis of syn-aminoalcohols via contributing
CH-p interaction: simple synthesis of (−)-bestatin
Byong Won Lee,a Jin Hwan Lee,a Ki Chang Jang,a Jae Eun Kang,a Jin Hyo Kim,a Ki-Min Parkb
and Ki Hun Parka,*
aDivision of Applied Life Science (BK21 program), Department of Agricultural Chemistry, Gyeongsang National University,
Chinju 660-701, South Korea
bThe Research Institute of Natural Science, Gyeongsang National University, Chinju 660-701, South Korea
Received 16 April 2003; revised 29 May 2003; accepted 30 May 2003
Abstract—1H NMR and X-ray crystallography studies revealed that a CH-p and chelation control in aromatic aminoaldehydes
(1–6) effects a highly diastereoselective addition to afford optically active syn-aminoalcohols (1a–6a). This methodology was
applied to the synthesis of (−)-bestatin.
© 2003 Elsevier Ltd. All rights reserved.
The stereoselective transformation of aromatic a-
aminoaldehydes to syn-aminoalcohols is attractive syn-
thetic strategy, since aromatic syn-a-hydroxy-b-amino
acid is a key unit of small peptides (e.g. bestatin,
phebestin and probestin) which function as aminopepti-
dase inhibitor.1 In the stereoselective transformation of
a-aminoaldehydes to aminoalcohols using organometal-
lic reagent, the most frequently used protecting group is
benzyl (Bn) group. The benzyl group allowed a high
degree of diastereoselectivity in nucleophilic addition of
a-aminoaldehydes, but produced an anti-aminoalcohols
with non-chelating control.2 Thus, aromatic a-amino
acids have been modified by a number of methods3
such as asymmetric dihydroxylation,4 Ojimas’ ring
opening procedure,5 etc. Our previous work has
demonstrated that a diastereoselective addition utilizing
the 9-phenylfluoren-9-yl (Pf) group affords syn-
aminoalcohol.6 The extension and application of this
methodology for the asymmetric synthesis of a syn-a-
hydroxy-b-amino acid is described herein. We also
described the first evidence for a CH-p interaction to
contribute a highly diastereoselective addition using
NMR technique and X-ray study.
The requisite substrates (1–10) were prepared easily by
the usual method from commercially available phenyl-
alanine, p-nitrophenylalanine, tyrosine, valine, leucine,
alanine, and serine. We chose the 9-phenylfluoren-9-yl
(Pf) group for protection of amine since this protecting
group has been shown to inhibit deprotonation at the
a-position of a-aminoaldehyde.7 a-Aminoaldehydes
having the Pf group are stable to Grignard reaction
condition.8
(R)-Phenylalaninal 1 was treated with ethynylmagne-
sium bromide at −40°C for 10 min to give syn-aminoal-
cohol 1a as a 9.5:1 ratio in 96% yield via a
diastereoselective addition (vida infra). As shown in
Table 1, other aromatic a-aminoaldehydes (2–6) were
also exposed to the same reaction conditions to give syn
products (2a–6a) in high selectivity and quantitative
yields. While, treatments of aliphatic a-aminoaldehydes
(7–10) under same condition afforded a less than 3:1
ratio of the syn and anti isomers. The above results
show that aromatic aminoaldehydes (1–6) provide
much higher selectivity than aliphatic aminoaldehydes
(7–10). Each diastereomeric aminoalcohol given by the
Grignard reaction could easily be isolated in pure form
by column chromatography. The structures of all
aminoalcohols (1a/b–10a/b) were confirmed by their
characteristic spectroscopic data. The relative stereo-
chemistries of the products 1a and 1b were determined
Keywords: bestatin; AHPBA; syn-aminoalcohol; CH-p interaction;
a-hydroxy-b-amino acid.
* Corresponding author. Tel: +82-55-7515472; fax: +82-55-7570178;
e-mail: khpark@gshp.gsnu.ac.kr
0040-4039/$ - see front matter © 2003 Elsevier Ltd. All rights reserved.
doi:10.1016/S0040-4039(03)01394-7

5906
B. W. Lee et al. / Tetrahedron Letters 44 (2003) 5905–5907
from their 1H NMR spectra based on the coupling
constant values and 2D NOE experiments.†
Table 1. Diastereoselective addition of ethynylmagnesium
bromide to aminoaldehyde a
Figure 1. Transition state of diastereoselective addition.
Based on the chemical shifts in 1H NMR spectra,
upfield shifts of aromatic region were observed at aro-
matic a-aminoaldehydes (1–6) which showed a high
selectivity, but were not observed at aliphatic
aminoaldehydes (7–10). As shown in Figure 2, obvious
upfield shift12 appeared at only (d) and expected a
CH-p interaction between Pf and nitrophenyl groups.
This upfield shift should be caused by a CH-p
interaction13 between both aromatic groups. Further-
b
Compound
R
Ratio
Yield
(%) d
a/b (syn/anti) c
1
2
3
4
5
6
7
8
9
Benzyl (R) e
1a/1b (9.5/1)
2a/2b (9.7/1)
3a/3b (9.5/1)
4a/4b (10/1)
5a/5b (10/1)
6a/6b (10/1)
7a/7b (1.8/1)
8a/8b (2.5/1)
9a/9b (2.2/1)
10a/10b (1.5/1)
96
97
94
96
97
97
94
95
93
94
Benzyl (S) f
4-Methoxybenzyl (R)
4-Methoxybenzyl (S)
4-Nitrobenzyl (R)
4-Nitrobenzyl (S)
2-Propyl (S)
iso-Butyl (S)
Methyl (S)
Benzyloxymethyl (S)
more, the structure of Pf- -Phe-OCH3 12 was con-
L
firmed by X-ray crystallography‡ as shown in Figure 3.
In the crystal structure, the distances which could be
considered as typical intramolecular CH-p interactions
10
14
,
(3.13–3.45 A) between benzyl (C20ꢀC25) and Pf
a All aminoaldehydes are enantiomeric pure.
,
group (C2ꢀC7) were observed: H3···centroid 3.00 A,
b The ratio was defined by 1H NMR spectrum of crude mixture.
c The stereochemistry was assigned by 2D-NOE experiment.
d Isolated yield.
ÚC3ꢀH3···centroid 117.0°, centroid···centroid 4.58 A.
,
This CH-p interaction may contribute to chelation con-
trol and a high selectivity.
e R-configuration.
f S-configuration.
Reetz reported that anti-aminoalcohols were obtained
from nucleophilic addition of a-aminoaldehydes pro-
tected with Bn group through non-chelating control,
due to the steric effect of Bn group.9 Even though
anti-aminoalcohols were expected from a-aminoalde-
hydes protected with Pf because of same reason, syn-
aminoalcohols were obtained in this study. The
syn-diastereoselectivity observed in conversion of aro-
matic aminoaldehydes (1–6) to aminoalcohols (1a–6a)
may rationalize that the chelation controlled Cram
cyclic model10 is more favorable than the Felkin–Anh
transition state model11 as shown in Figure 1. Why
could there not be observed a high selectivity in
aliphatic a-aminoaldehydes?
Figure 2. Aromatic region in 1H NMR (500 MHz, CDCl3).
(a) N-Pf-Leu-OCH3; (b) p-NO2-Phe-OCH3; (c) N-Boc-p-
NO2-Phe-OCH3; (d) N-Pf-p-NO2-Phe-OCH3.
† The compounds 1a and 1b were cyclized with carbonyldiimidazole
after hydrogenation in Pd/H2 condition to give 11a and 11b,
respectively. Based on the coupling constant values in the H NMR
‡ Crystal data for 12; C29H25NO2 M=419.50, orthorhombic, space
1
,
group P212121, a=7.9757(6), b=15.4702(12), c=18.4058(14) A,
3
U=2271.0(3) A , Z=4, v=0.076 mm−1, Rint=0.0326. A total of
spectra of cyclized syn and anti isomers as like 1a/1b, the stereo-
,
chemistries of syn isomer (J3,4=11–13 Hz) and anti isomers (J3,4
=
14347 reflections were measured for the angle range 3.44<2q<56.58
and 5294 independent reflections were used in the refinement. The
final parameters were wR2=0.0871 and R1=0.0398 [I>2|(I)]. Data
were collected on a Bruker SMART diffractometer equipped with a
6–8 Hz) could be determined from each coupling constant values.
,
graphite monochromated Mo Ka (u=0.71073 A) radiation source
and a CCD detector using the SAINT-NT software.15 The struc-
tures were solved by direct methods and refined by full matrix
least-squares against F2 for all data using the SHELXTL program
package.15 CCDC reference number 200283.

B. W. Lee et al. / Tetrahedron Letters 44 (2003) 5905–5907
5907
To summarize the above results, we developed an
efficient method for the preparation of syn-aminoalco-
hol (1a–6a) from readily available aromatic a-amino
acids through chelation control contributed by a CH-p
interaction. These species can be easily converted to
each corresponding syn-a-hydroxy-b-amino acids as
key component of small peptide. We have also applied
1a to enantiomerically pure (−)-bestatin.
Acknowledgements
This work was supported by grant (No R01-2000-
00175) from Basic Research Program of Korea Science
& Engineering Foundation. We are also grateful for the
financial support of Brain Korea 21 program.
References
Figure 3. Molecular structure and atomic numbering scheme
of 12. Thermal ellipsoids are drawn at the 30% probability
level and the H-atoms omitted for clarity. The dihedral angle
between the aromatic planes of benzyl and Pf group
1. (a) Nagai, M.; Kojima, F.; Naganawa, H.; Hamada, M.;
Aoyagi, T.; Takeuchi, T. J. Antibiot. 1997, 50, 82; (b)
Aoyagi, T.; Yoshida, S.; Nakamura, Y.; Shigihara, Y.;
Hamada, M.; Takeuchi, T. J. Antibiot. 1990, 43, 143; (c)
Umezawa, H.; Aoyagi, T.; Suda, H.; Hamada, M.;
Takeuchi, T. J. Antibiot. 1976, 29, 97.
,
(C(2)ꢀC(7)) is 34.41(5)°. Selected H···C distances (A):
H(3)···C(20) 3.22, H(3)···C(21) 3.45, H(3)···C(22) 3.44,
H(3)···C(23) 3.29, H(3)···C(24) 3.13, H(3)···C(25) 3.16.
2. Shibata, N.; Itoh, E.; Terashima, S. Chem Pharm. Bull.
1998, 46, 733.
This methodology has been applied to the synthesis of
3. (a) Hagihara, M.; Schreiber, S. L. J. Am. Chem. Soc.
1992, 114, 6570; (b) Dondoni, A.; Perrone, D.; Semola,
T. Synthesis 1995, 181; (c) Nozaki, K.; Sato, N.; Takaya,
H. Tetrahedron: Asymmetry 1993, 4, 2179.
(−)-bestatin 16 which requires the coupling of two
structural
unit,
N-terminal
b-amino-a-hydroxy
phenylbutanoic acid (AHPBA) and C-terminal amino
acid leucine. After protection of hydroxy group in 1a
with BnBr, resulting benzylate was oxidized with
KMnO4 to give AHPBA derivative 14 as N-terminal
part in 84% yield. The coupling reaction of 14 with
(S)-leucine methyl ester was carried out in the presence
of DCC to afford the dipeptide product 15 in 91%
yield. Treatment of 15 with lithium hydroxide followed
by exposure of resulting crude acid to H2 over Pd/C
gave (−)-bestatin in 93% yield, which was identical with
a previous report (Scheme 1).1
4. Wang, Z.-M.; Kolb, H. C.; Sharpless, K. B. J. Org.
Chem. 1994, 59, 5104.
5. Ojima, I.; Delaloge, F. Chem. Soc. Rev. 1997, 26, 377.
6. (a) Lee, W. S.; Jang, K. C.; Kim, J. H.; Park, K. H.
Chem. Commun. 1999, 251; (b) Chang, K.-T.; Park, H.-
Y.; Kim, Y.-K.; Park, K. H.; Lee, W. S. Heterocycles
2001, 55, 1173.
7. Lubell, W. D.; Rapoport, H. J. Am. Chem. Soc. 1987,
109, 236.
8. Gerspacher, M.; Rapoport, H. J. Org. Chem. 1991, 56,
3700.
9. Reetz, M. T.; Drewes, M. W.; Schmitz, A. Angew. Chem.,
Int. Ed. Engl. 1987, 26, 1141.
10. Garner, P.; Ramakanth, S. J. Org. Chem. 1986, 51, 2609.
11. Anh, N. T.; Eisenstein, O. Nouv. J. Chem. 1977, 1, 61.
12. Shetty, A. S.; Zhang, J.-S.; Moore, J. S. J. Am. Chem.
Soc. 1996, 118, 1019.
13. Wakita, R.; Miyakoshi, M.; Nakatsuji, Y.; Okahara, M.
J. Inclu. Phenom. 1991, 10, 127.
14. (a) Hunter, C. A.; Sanders, J. K. M. J. Am. Chem. Soc.
1990, 112, 5525; (b) Hunter, C. A.; Singh, J.; Thornton, J.
M. J. Mol. Biol. 1991, 218, 837; (c) Singh, J.; Thornton,
J. M. J. Mol. Biol. 1990, 211, 595.
15. SAINT-NT version 5.0. Program for data collection and
data reduction, Bruker-AXS, Madison, WI, 1998;
Sheldrick, G. M. SHELXTL version 5.1. Program for
Solution and Refinement of Crystal Structures, Bruker-
AXS, Madison, WI, 1998.
Scheme 1. Reagents and conditions: (i) BnBr, NaH, Bu4NI,
THF, 0°C, 97%; (ii) KMnO4, HOAc, H2O, pentane, 87%; (iii)
L
-Leu-OCH3, DCC, HOBT, TsOH, Et3N, THF, 0°C, 91%;
(iv) LiOH, THF/H2O, 0°C, 95%; (v) H2, Pd/C, MeOH, 50°C,
93%.